Researchers at our lab found satisfaction increased and prodcutivity doubled in programming teams that moved from offices to an open-plan office. Workers also had access to an unassigned outside office for private phone calls, deep-think problems, etc. This study didn't include cubicles, although I tend to agree with OxB that cubes are the worst of both options without the benefits.
The reference is: Teasley, S.D., Covi, L., Krishnan, M.S., & Olson, J.S. (2000). How does radical collocation help a team succeed? In Proceedings of CSCW 2000 (pp. 339-346). New York: ACM Press.
Contact me or the authors and we can send you a pdf. See also the technical reports link from the CREW web page at crew.umich.edu
I'd think this would be a lousy way to index commercial sites, which would avoid linking to each other. e.g. the one link you're sure not to find on Ford.com is Generalmotors.com!
Yeah, the demise of Hypercard has been a real low point in the history of Macintosh. HC used to be the best amateur-hacker environment around, and it also attracted a lot of people to the Mac. The Mac is not the same without it. It seems like HC fell out of favor just when the Web came around-- maybe because HC failed to very integrate with the Web, or maybe because html offered another easy outlet for part-time geeks.
Correlation vs causality. Actually this study has a pretty strong claim for causation, because the second study is experimental. As all stats 101 students know, you can't assume causation from a correlation in survey-type study. A survey (study 1) leaves you vulnerable to the argument "maybe its the violent people who choose to play video games", or "maybe a 3rd variable, like mental illness, causes both." However, when you do a controlled experimental study (as they did) you know what order events happen in time. FIRST, they brought in a group of students and controlled for various attributes. THEN they had them play two diff games. THEN they observed the ones who played the violent game to behave more aggresively. Doesn't PROVE causation but it's pretty strong evidence for it. SECOND, they never claim that video games turn a non-violent person into an aggressive one. They claim that it is one factor among several that primes aggressive behavior. So, take a frustrated and confused teen, add years of mimicking violent behavior on the screen, and you're more likely to get real violence. Is that important? Hell, yes it is! As someone with children nearing school age, I'm in favor of controlling every variable that it is possible to control to minimize violence.
Faster brains not faster machines
on
Faster
·
· Score: 2
I started reading Hemos' review prepared to argue: faster machines don't make as much difference as faster brains & people. Looks like Gleick appears to already agree with me. Spoils a good argument. I'll make it anyway because my desktop is busy processing log files. Argument: more information doesn't speed up the pce of life, because available info has far outstripped available mental resources for a long time. Question: when was the last time anyone could be a true 'renaissance man (or woman' and have a handle on everthing going on in science, politics, economics, etc? Answer: the renaissance. Even in the 1950's nobody could really keep up with everything going on around them. The speed of life is determined by the speed at which people are able to, and are willing to process info and make decisions. The speed of life picks up when more and more people are willing to work longer hours, multi-task, multi-think, multi-multi. So why are things faster now? because culture has changed, and because the avg IQ of the population has been increasing steadily over the years. No one knows why. Probably IQ is increased by a generally stimulating environment more than by formal schooling, which hasn't improved much.
The lack of any kind of business model makes ArsDigita University not very interesting, I fear. It's pure philanthropy-- Greenspun's company is footing the bill as an act of pure public service. Service that only benefits 30 students per yr with high intelligence and SATs. Is this the most worthy cause he could find? This model is not sustainable, and won't be widely copied. What's the point? What I'd like to see is true open source university model, that combines money from Universities, corporations, alumni, advertising dollars, whatever, to provide free high-tech teaching. How's this for a model: companies donate employee time to teach current software classes. In return they get access to graduates as interns and new employees. Universities donate faculty time to teach the more basic-science courses on data structures, algorithms, mathematics, etc. In return they get... well I haven't figured that part out yet. Maybe there's a model where graduates who land high-paying gigs give back to the University, and the more they make the more they're obligated to give? Although I think that's been tried (at Yale?) with only moderate success... Any ideas?
I worked in a school where the science dept had laptops at aprox a ratio of 1 laptop per 2 kids. But after a couple of years they had enough trouble with theft, breakage, etc. that they cabled each laptop to the desks. They still uncabled them for a few important purposes. 1) students could check them out to do homework on, and 2) they would do in-the-field data collection at local streams, etc. But for the most part, they'd have been better off with a bunch of (cheaper, faster, more ergonomic) desktop machines and had a bank of laptops avaailable for checkout.
I disagree that computers tend to distract students from deep principles. When I tutor 9th grade algebra I am amazed at how much of the students' cognitive capacity is wrapped up in the mechanics of the paper-and-pencil algebra (anyone else remember binomial Benny?) and how little is left over for thinking about the problem. Ideally, what computers can do is routinize the stuff that is trivial and allow students to focus on what's central and important. If entire classes of students are focused on the mechanics of using a particular software package and not on the underlying principles, that has a simple explanation: bad teaching. And that happens both with and without computers.
Slashdot Mirror
The reference is: Teasley, S.D., Covi, L., Krishnan, M.S., & Olson, J.S. (2000). How does radical collocation help a team succeed? In Proceedings of CSCW 2000 (pp. 339-346). New York: ACM Press.
Contact me or the authors and we can send you a pdf. See also the technical reports link from the CREW web page at crew.umich.edu
I'd think this would be a lousy way to index commercial sites, which would avoid linking to each other. e.g. the one link you're sure not to find on Ford.com is Generalmotors.com!
Yeah, the demise of Hypercard has been a real low point in the history of Macintosh. HC used to be the best amateur-hacker environment around, and it also attracted a lot of people to the Mac. The Mac is not the same without it. It seems like HC fell out of favor just when the Web came around-- maybe because HC failed to very integrate with the Web, or maybe because html offered another easy outlet for part-time geeks.
Correlation vs causality. Actually this study has a pretty strong claim for causation, because the second study is experimental. As all stats 101 students know, you can't assume causation from a correlation in survey-type study. A survey (study 1) leaves you vulnerable to the argument "maybe its the violent people who choose to play video games", or "maybe a 3rd variable, like mental illness, causes both." However, when you do a controlled experimental study (as they did) you know what order events happen in time. FIRST, they brought in a group of students and controlled for various attributes. THEN they had them play two diff games. THEN they observed the ones who played the violent game to behave more aggresively. Doesn't PROVE causation but it's pretty strong evidence for it. SECOND, they never claim that video games turn a non-violent person into an aggressive one. They claim that it is one factor among several that primes aggressive behavior. So, take a frustrated and confused teen, add years of mimicking violent behavior on the screen, and you're more likely to get real violence. Is that important? Hell, yes it is! As someone with children nearing school age, I'm in favor of controlling every variable that it is possible to control to minimize violence.
I started reading Hemos' review prepared to argue: faster machines don't make as much difference as faster brains & people. Looks like Gleick appears to already agree with me. Spoils a good argument. I'll make it anyway because my desktop is busy processing log files. Argument: more information doesn't speed up the pce of life, because available info has far outstripped available mental resources for a long time. Question: when was the last time anyone could be a true 'renaissance man (or woman' and have a handle on everthing going on in science, politics, economics, etc? Answer: the renaissance. Even in the 1950's nobody could really keep up with everything going on around them. The speed of life is determined by the speed at which people are able to, and are willing to process info and make decisions. The speed of life picks up when more and more people are willing to work longer hours, multi-task, multi-think, multi-multi. So why are things faster now? because culture has changed, and because the avg IQ of the population has been increasing steadily over the years. No one knows why. Probably IQ is increased by a generally stimulating environment more than by formal schooling, which hasn't improved much.
The lack of any kind of business model makes ArsDigita University not very interesting, I fear. It's pure philanthropy-- Greenspun's company is footing the bill as an act of pure public service. Service that only benefits 30 students per yr with high intelligence and SATs. Is this the most worthy cause he could find? This model is not sustainable, and won't be widely copied. What's the point? What I'd like to see is true open source university model, that combines money from Universities, corporations, alumni, advertising dollars, whatever, to provide free high-tech teaching. How's this for a model: companies donate employee time to teach current software classes. In return they get access to graduates as interns and new employees. Universities donate faculty time to teach the more basic-science courses on data structures, algorithms, mathematics, etc. In return they get... well I haven't figured that part out yet. Maybe there's a model where graduates who land high-paying gigs give back to the University, and the more they make the more they're obligated to give? Although I think that's been tried (at Yale?) with only moderate success... Any ideas?
I worked in a school where the science dept had laptops at aprox a ratio of 1 laptop per 2 kids. But after a couple of years they had enough trouble with theft, breakage, etc. that they cabled each laptop to the desks. They still uncabled them for a few important purposes. 1) students could check them out to do homework on, and 2) they would do in-the-field data collection at local streams, etc. But for the most part, they'd have been better off with a bunch of (cheaper, faster, more ergonomic) desktop machines and had a bank of laptops avaailable for checkout.
I disagree that computers tend to distract students from deep principles. When I tutor 9th grade algebra I am amazed at how much of the students' cognitive capacity is wrapped up in the mechanics of the paper-and-pencil algebra (anyone else remember binomial Benny?) and how little is left over for thinking about the problem. Ideally, what computers can do is routinize the stuff that is trivial and allow students to focus on what's central and important. If entire classes of students are focused on the mechanics of using a particular software package and not on the underlying principles, that has a simple explanation: bad teaching. And that happens both with and without computers.